SynthTrail

Understand how RAG crawlers retrieve your content

Retrieval systems fetch pages to answer a question, then stop. Their trails reveal what they chose and how they chose it. SynthTrail records those trails for browser-based retrievers.

Install the snippet See a sample trail

$9.99 a month. No contract. Cancel any time.

The problem: retrieval is a black box

You cannot see why an assistant picked your page, what else it read first or how long it stayed. Without that, improving your pages for retrieval is guesswork.

What the trails reveal

The sequence of pages, the gap between them, whether there is any scrolling, and whether the session ends after a single fetch or goes deep. Short, targeted sessions look like retrieval; long, wide ones look like collection.

Use what you learn

Compare the pages retrievers open most with the pages you want them to use, then change structure, titles or summaries and watch the next sessions. Retrievers that fetch HTML without running scripts will not appear, so your picture is partial.

What you can learn

Which pages assistants choose
Order and depth of retrieval
Short lookups versus long sweeps
Change over time as you edit your pages

Follow the trail on your own site

Paste one snippet and see the synthetic visitors that use your site. Card details are entered on Stripe, never on this site.

Install the snippet